Intelligent Verification System and Method for Railway Tank Cars

Abstract

This application relates to the field of automated loading and unloading technology for railway oil transportation, and provides an intelligent verification system and method for railway tank cars. The method involves acquiring relevant data on incoming oil, obtaining the tank car's positioning status information through an identification device, and collecting the tank car number image using an image recognition device and converting it into character information. The character information is then compared with the corresponding batch of tank car number information in the refinery's incoming oil information database. If the comparison passes, metering data acquisition is initiated, and metering data of the oil inside the tank car is acquired using a portable measuring instrument. The collected metering data is compared and analyzed with theoretical data from the refinery's incoming oil information database and historical data from the railway tank car metering database using a preset intelligent algorithm. If the comparison passes, verification is performed based on the car number, positioning status, and metering data. Upon successful verification, an automatic unloading command is generated. The railway oil unloading automation control system receives the automatic unloading command and completes the intelligent verification of the railway tank car.

Description

Technical Field

[0001] This application relates to the field of automated loading and unloading technology for railway oil transportation, and in particular to an intelligent verification system and method for railway tank cars. Background Technology

[0002] In the railway oil loading and unloading operations of the petrochemical industry, the verification process for railway tank cars typically includes key steps such as car number verification, placement status confirmation, metering data verification, and unloading operation control. Traditionally, these processes rely heavily on manual operation and experience-based judgment. For example, visually verifying car numbers is prone to errors due to factors like lighting and angle; placement status confirmation depends on operators checking the relative positions of the tank car and the loading / unloading port, which introduces subjective judgment errors; metering data verification relies solely on single data collection using portable devices and simple comparison with theoretical values, without incorporating historical data for trend analysis, making it difficult to detect equipment anomalies or potential data fluctuations; and the confirmation of the start and end status of unloading operations relies on manual inspections, which carries risks such as delayed response and safety interlock failures. Furthermore, existing technologies operate independently, lacking systematic intelligent integration, resulting in low verification efficiency, high human intervention costs, and poor data traceability, failing to meet the automated and intelligent operational requirements of modern refineries.

[0003] While existing technologies offer some improvements for single stages (such as standalone vehicle number recognition systems or metering data acquisition devices), no solution yet organically combines image recognition, multi-source data fusion (refinery incoming oil information database, railway tank car metering database), intelligent algorithm comparison, and automated control processes to form a complete intelligent verification system covering "positioning status recognition - vehicle number verification - intelligent analysis of metering data - unloading command generation - full-process operation control." In particular, existing technologies do not involve multi-dimensional comparative analysis of real-time metering data, theoretical data, and historical data using preset intelligent algorithms, nor do they disclose multi-level verification mechanisms and automated command generation logic based on vehicle number, positioning status, and metering data. Therefore, an innovative solution that can improve verification accuracy, automation, and operational safety is urgently needed. Summary of the Invention

[0004] This application provides an intelligent verification system and method for railway tank cars, aiming to solve the problem that the existing technology does not involve multi-dimensional comparison and analysis of real-time measurement data, theoretical data and historical data through preset intelligent algorithms, nor does it disclose a multi-level verification mechanism and automated instruction generation logic based on car number, positioning status and measurement data.

[0005] Firstly, this application provides an intelligent verification method for railway tank cars, including: After obtaining relevant data on incoming oil, the system uses identification equipment to obtain the positioning status information of the tanker truck, and uses image recognition equipment to collect the vehicle number image of the tanker truck and convert it into character information. Compare the character information with the tanker truck number information of the corresponding batch in the refinery's oil arrival information database: if the comparison is successful, start the metering data acquisition and obtain the oil metering data in the tanker truck through a portable measuring instrument; The collected metering data is compared and analyzed with theoretical data from the refinery's oil arrival information database and historical data from the railway tank car metering database using a preset intelligent algorithm. If the comparison passes, the data is verified based on the car number, entry status, and metering data. Once the verification passes, an automatic oil unloading command is generated. The railway oil unloading automation control system receives the automatic oil unloading command, confirms the start of the oil unloading status, starts the oil unloading operation after confirming that the equipment status is normal, confirms the end status after the oil unloading is completed, generates an oil receipt and dispatch certificate information report, and completes the intelligent verification of railway tank cars.

[0006] Secondly, this application provides an intelligent verification system for railway tank cars, comprising: The data acquisition unit is used to acquire relevant data on incoming oil, obtain the positioning status information of the oil tanker through the recognition device, and collect the vehicle number image of the oil tanker and convert it into character information according to the image recognition device. The information comparison unit is used to compare the character information with the tanker truck number information of the corresponding batch in the refinery's oil arrival information database. If the comparison is successful, the metering data acquisition is started, and the metering data of the oil in the tanker is obtained through a portable measuring instrument. The instruction generation unit is used to compare and analyze the collected metering data with the theoretical data of the refinery's oil arrival information database and the historical data of the railway tank car metering database using a preset intelligent algorithm. If the comparison is successful, the unit verifies the data based on the car number, the entry status, and the metering data. Once the verification is successful, an automatic oil unloading instruction is generated. The intelligent verification unit is used by the railway oil unloading automation control system to receive automatic oil unloading commands, confirm the start status of oil unloading, start the oil unloading operation after confirming that the equipment status is normal, confirm the end status after oil unloading is completed, generate an oil receipt and dispatch certificate information report, and complete the intelligent verification of railway tank cars.

[0007] Thirdly, this application also provides a computer device, comprising: Memory and processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, when executing the computer program, implement the steps of the intelligent verification method for railway tank cars as described in the first aspect above.

[0008] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the intelligent verification method for railway tank cars as described in the first aspect above.

[0009] This application provides an intelligent verification system and method for railway tank cars. The method automatically acquires the tank car's positioning status and car number information through identification equipment, replacing manual visual inspection and manual data entry, reducing human error and improving operational efficiency. By comparing and analyzing real-time metering data with refinery theoretical data and historical operational data through a preset intelligent algorithm, combined with logical verification rules, abnormal data is automatically screened, improving data verification accuracy compared to traditional single data comparison. Multiple confirmations of equipment status are performed before unloading operations, and status changes are monitored in real time during unloading. After unloading, an unalterable report record is automatically generated, constructing a safety protection system covering the entire operation cycle and reducing risks such as leaks and static electricity hazards caused by human negligence. By storing metering data on blockchain, automatically generating and encrypting oil receipt and dispatch certificate reports, the entire process of operational data is traceable, meeting the digital management and compliance audit needs of refining and chemical enterprises.

[0010] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic flowchart illustrating the steps of an intelligent verification method for railway tank cars provided in an embodiment of this application; Figure 2 This is a logical schematic diagram of an intelligent verification method for railway tank cars provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an intelligent verification system for railway tank cars provided in one embodiment of this application; Figure 4 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0013] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0016] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0017] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0018] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0019] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0020] In the railway oil loading and unloading operations of the petrochemical industry, the verification process for railway tank cars typically includes key steps such as car number verification, placement status confirmation, metering data verification, and unloading operation control. Traditionally, these processes rely heavily on manual operation and experience-based judgment. For example, visually verifying car numbers is prone to errors due to factors like lighting and angle; placement status confirmation depends on operators checking the relative positions of the tank car and the loading / unloading port, which introduces subjective judgment errors; metering data verification relies solely on single data collection using portable devices and simple comparison with theoretical values, without incorporating historical data for trend analysis, making it difficult to detect equipment anomalies or potential data fluctuations; and the confirmation of the start and end status of unloading operations relies on manual inspections, which carries risks such as delayed response and safety interlock failures. Furthermore, existing technologies operate independently, lacking systematic intelligent integration, resulting in low verification efficiency, high human intervention costs, and poor data traceability, failing to meet the automated and intelligent operational requirements of modern refineries.

[0021] While existing technologies offer some improvements for single stages (such as standalone vehicle number recognition systems or metering data acquisition devices), no solution yet organically combines image recognition, multi-source data fusion (refinery incoming oil information database, railway tank car metering database), intelligent algorithm comparison, and automated control processes to form a complete intelligent verification system covering "positioning status recognition - vehicle number verification - intelligent analysis of metering data - unloading command generation - full-process operation control." In particular, existing technologies do not involve multi-dimensional comparative analysis of real-time metering data, theoretical data, and historical data using preset intelligent algorithms, nor do they disclose multi-level verification mechanisms and automated command generation logic based on vehicle number, positioning status, and metering data. Therefore, an innovative solution that can improve verification accuracy, automation, and operational safety is urgently needed.

[0022] To resolve the above issues, please refer to [link / reference]. Figure 1 , Figure 1 This is a schematic flowchart of an embodiment of the intelligent verification method for railway tank cars provided in this application. This intelligent verification method for railway tank cars can be implemented by computer equipment, which can be deployed on a single server or a server cluster. It can also be deployed on handheld terminals, laptops, wearable devices, or robots, etc.

[0023] It should be noted that the acquisition of any information involved in the provided methods is in compliance with relevant regulations and is carried out with the user's consent, and will not infringe on the user's privacy or violate relevant laws and regulations.

[0024] Specifically, such as Figure 1 As shown, the provided intelligent verification method for railway tank cars includes steps S101 to S104, which are detailed below: Step S101. After obtaining the relevant data of the incoming oil, obtain the in-position status information of the tanker through the identification device, and collect the vehicle number image of the tanker through the image recognition device and convert it into character information.

[0025] Specifically, collect the in-position status and vehicle number information of the tanker through the multi-modal device to provide basic data for subsequent verification. Specifically, it includes: Obtaining relevant incoming oil data: Obtain batch information in the refinery incoming oil plan (such as oil product type, planned vehicle number, estimated weight / volume, loading and unloading port number, etc.), tanker basic information (vehicle type, load limit, historical metering data index, etc.). In-position status information collection: Identify whether the tanker is accurately parked at the designated loading and unloading position, including position alignment (lateral / longitudinal deviation between the tanker and the loading and unloading port), connection status (whether the loading and unloading arm / crane pipe is aligned with the interface), safety status (whether the anti-rolling device is in place, whether the static electricity grounding is connected), etc. Vehicle number image recognition and character conversion: Collect the vehicle number area image through the image recognition device, and use optical character recognition (OCR) technology to convert it into machine-readable character information.

[0026] The data acquisition interface docks with the refinery ERP system or the incoming oil scheduling platform through the API to synchronize data such as the vehicle number list, loading and unloading port allocation, and theoretical metering value of the incoming oil batch in real time.

[0027] The in-position status detection includes: Hardware deployment: Install laser range sensors (to detect lateral / longitudinal position deviation), industrial cameras (to photograph the alignment status of the loading and unloading port and the tanker interface), pressure sensors (to detect whether the loading and unloading arm connection is in place), and wireless sensors (to monitor the status of the anti-rolling device and static electricity grounding) at the loading and unloading platform. Data fusion: Input the multi-sensor data into the edge computing module, and judge the in-position status through preset rules (such as deviation < 5 cm, loading and unloading arm pressure > threshold, static electricity grounding signal normal), and output the status of "qualified in-position" or "abnormal in-position".

[0028] Vehicle number recognition includes: Image acquisition: Install a high-definition camera (equipped with a fill light to adapt to different lighting conditions) above the tanker's travel route, trigger the camera to take a photo after the tanker stops, and collect the vehicle number area image (supporting side / top vehicle number recognition to cover different perspectives). Character conversion: Adopt a deep learning OCR model (such as the CRNN+CTC architecture), preprocess the image (denoising, perspective correction), then recognize the characters, output the vehicle number string (such as "Tank A-12345"), and mark the confidence level (such as ≥95% is regarded as valid recognition).

[0029] Step S102. Compare the character information with the tanker vehicle number information of the corresponding batch in the refinery incoming oil information database: If the comparison passes, start the metering data collection, and obtain the oil product metering data in the tanker through a portable measuring instrument.

[0030] Specifically, the identified vehicle number is matched with the planned vehicle number in the refinery's oil arrival information database to verify the legality of the tanker truck's identity; if the comparison passes, the metering data collection process is triggered to obtain real-time oil metering parameters.

[0031] The vehicle number comparison and verification process precisely matches the character information generated in step S101 with the list of vehicle numbers corresponding to the oil batch. It supports fuzzy matching (such as allowing differences in case and spaces) and fault tolerance mechanisms (such as triggering manual review after 3 consecutive recognition failures).

[0032] The metering data acquisition is triggered remotely only when the vehicle number comparison passes, activating a portable measuring instrument (such as a smart level gauge, temperature sensor, or density meter) to collect real-time metering data of the oil (level height, temperature, density, volume / weight, etc.).

[0033] The vehicle number comparison logic includes: Data matching: The system performs a precise string match between the current batch of vehicle numbers in the future fuel information database (e.g., containing 10 vehicle numbers) and the recognition result, supporting "one vehicle, one comparison" (verifying each vehicle number individually) or "batch full matching" (verifying all arriving vehicle numbers in batches). Anomaly handling: If the recognized vehicle number is not in the planned list (e.g., confidence level ≥ 95% but no matching record), the system automatically marks it as "vehicle number abnormal," triggers an audible and visual alarm, and sends a manual review instruction (e.g., the on-site operator confirms the vehicle number via a handheld terminal).

[0034] The linkage of metering equipment includes: Equipment integration: The portable measuring instrument connects to the on-site industrial control computer via Bluetooth / Wi-Fi. After the industrial control computer receives the vehicle number and the comparison is successful, it sends a command to activate the equipment's data acquisition function (such as automatically lowering the level gauge probe into the tank truck). Data acquisition items: At least the following data must be collected: oil temperature (for volume conversion), liquid level (to calculate the current volume), density (to convert to standard volume), and the temperature difference between the oil and the tank wall (to determine if stratification exists). The data accuracy must meet industry standards (e.g., temperature ±0.5℃, liquid level ±1mm).

[0035] Step S103. The collected metering data is compared and analyzed with the theoretical data of the refinery's oil arrival information database and the historical data of the railway tank car metering database using a preset intelligent algorithm. If the comparison is successful, the verification is performed based on the car number, entry status and metering data. After successful verification, an automatic oil unloading command is generated.

[0036] Specifically, by using a pre-set intelligent algorithm, real-time metering data, refinery theoretical data (planned values), and tank truck historical data (historical metering records of the same vehicle model / number) are analyzed from multiple dimensions to build a multi-level verification mechanism; if the verification is successful, an automated unloading command is generated.

[0037] Multi-source data fusion analysis includes: Theoretical data comparison: comparing real-time measurement values ​​with theoretical values ​​(such as planned volume and density range) in the oil supply information database, and calculating absolute and relative deviations (such as volume deviation rate ≤ ±2%). Historical data trend analysis: calling up the measurement data of the last 3 transports of the railway tank car in the railway tank car measurement database, and analyzing the fluctuation trend of density and volume (such as triggering equipment anomaly warnings when two consecutive density drops > 5%).

[0038] Intelligent algorithm verification uses statistical models (such as the 3σ principle to identify outliers) or machine learning algorithms (such as isolated forests to detect outliers in data) to comprehensively judge whether the measurement data is reliable.

[0039] The multi-level verification mechanism includes: vehicle number legality (passing step S102), qualified entry status (judgment in step S101), and qualified metering data verification (analysis results of this step). All three must be satisfied simultaneously before an oil unloading command can be generated.

[0040] Data comparison rules include: Theoretical value verification: The calculation formula is deviation rate = (real-time value - theoretical value) / theoretical value × 100%. For example, a volume deviation rate ≤ ±2% and a density deviation rate ≤ ±1% are considered acceptable. If the deviation exceeds the range, it is marked as "measurement anomaly". Historical trend analysis: The density data from three historical transports of the same tanker truck are calculated using a moving average. If the deviation between the real-time density and the historical mean is > 3σ (standard deviation), and this occurs twice consecutively, it is judged as "potential equipment failure" (such as sensor drift).

[0041] The intelligent algorithm implementation includes: basic statistical model: preset reasonable ranges for density and temperature of various oil types (e.g., gasoline density 700-780 kg / m³). 3 (When real-time data exceeds the range, an early warning is triggered directly.)

[0042] Machine learning model: Train an anomaly detection model (such as One-Class SVM) based on historical normal operation data, input quantitative data vectors (volume, density, temperature, oil temperature difference) in real time, and output "normal" or "abnormal" classification results.

[0043] The instruction generation logic includes: generating an oil unloading instruction only when the following conditions are met simultaneously: vehicle number comparison status = "passed"; placement status = "qualified"; and measurement data verification result = "passed" (theoretical deviation is within the threshold and historical trend is not abnormal).

[0044] Step S104. The railway oil unloading automation control system receives the automatic oil unloading command, confirms the start status of oil unloading, starts the oil unloading operation after confirming that the equipment status is normal, confirms the end status after the oil unloading is completed, generates an oil receipt and dispatch certificate information report, and completes the intelligent verification of railway tank cars.

[0045] Specifically, the oil unloading command is sent to the automated control system to enable unmanned start and end confirmation of the oil unloading operation, while generating full-process data reports to ensure the traceability of the operation.

[0046] The oil unloading start confirmation verifies the status of the oil unloading pipeline valves and whether the safety interlock devices (such as emergency shut-off valves and combustible gas alarms) are functioning properly to prevent the risk of starting under pressure or leakage.

[0047] The operation process monitoring collects unloading flow and pressure data in real time to determine whether there are abnormal interruptions (such as the flow suddenly dropping to zero for more than 30 seconds) or excessive unloading (the cumulative amount exceeds the theoretical value by 110%).

[0048] After the oil unloading is completed, the remaining amount in the tanker truck is checked again (to prevent incomplete unloading), and an electronic report containing the truck number, time, metering data, and equipment status is generated and archived in the refinery's ERP system.

[0049] The integrated automation control includes: Hardware interface: Connecting to the unloading valve assembly, flow meter, and pressure sensor via Modbus / TCP protocol. Commands include specific operations such as "open valve X" and "start pump Y," supporting a dual confirmation mechanism (system command + field equipment feedback signal). Safety interlock: Before startup, it must be confirmed that the static grounding signal is normal, the anti-slip device is locked, and the flammable gas concentration is <20% of the lower explosive limit. If any condition is not met, the command will be refused and an alarm will be triggered.

[0050] Process monitoring and anomaly handling include: Real-time data acquisition: Collecting unloading flow rate and pipeline pressure at a frequency of 1 second / time, displaying curves in real time through the SCADA system, and setting flow fluctuation thresholds (e.g., triggering an early warning if instantaneous flow change > 20%). Anomaly response: If a leak alarm occurs (combustible gas concentration > threshold) or pressure anomaly, the system automatically triggers the emergency shut-off valve, stops operations, and pushes a fault code to the operation and maintenance platform.

[0051] Report generation and traceability include: Electronic document generation: including the "Railway Tank Car Acceptance Form" and "Oil Product Receipt and Dispatch Measurement Form," automatically filling in car number, operation time (start / end), measurement data (original dispatch volume, actual receipt volume, loss rate), and equipment number (such as the sensor ID involved in the measurement). Data archiving involves encrypting and storing reports and process data (images, sensor logs, instruction records) in a blockchain database or distributed file system, supporting fast retrieval by car number, time, and batch, meeting audit requirements.

[0052] In some embodiments, obtaining the positioning status information of the oil tanker through the identification device includes: deploying a laser ranging sensor, a pressure sensor, and a visual recognition device at a preset position on the track in the railway unloading operation area; collecting distance data between the oil tanker and the loading / unloading port through the laser ranging sensor; collecting wheel-to-track pressure distribution data through the pressure sensor; and collecting the relative position image of the coupler and the track marking line through the visual recognition device; inputting the distance data, pressure distribution data, and position image into an edge computing unit; performing logical judgment based on a preset positioning rule library; and determining that the oil tanker is ready to be positioned if the distance data is within a safe operating range, the pressure distribution data meets the track load balance condition, and the deviation between the coupler and the marking line in the position image is less than a preset threshold.

[0053] By fusing multi-source data from laser ranging, pressure sensing, and visual recognition, the system accurately determines whether the tanker truck is correctly positioned. The core components include: Multi-dimensional data acquisition: Laser ranging sensors acquire the spatial distance between the tanker truck and the loading / unloading port; pressure sensors monitor the uniformity of pressure distribution between the wheels and the rails; and visual recognition equipment captures the relative position of the coupler and the rail markings to construct a three-dimensional spatial positioning model. Edge computing logic judgment: The three types of data are input into the edge computing unit, and logical verification is performed based on a preset rule base (safety distance, pressure balance threshold, and position deviation threshold) to ensure that the tanker truck's parking position meets the safety and alignment accuracy requirements of the loading / unloading operation.

[0054] The hardware deployment includes: Laser rangefinders: Two sets (one horizontal and one vertical) are installed on brackets on both sides of the loading / unloading port. The horizontal sensor is perpendicular to the track and measures the horizontal distance between the side of the tank car and the loading / unloading port (accuracy ±2mm); the vertical sensor is parallel to the track and measures the longitudinal distance between the end face of the tank car and the loading / unloading port (accuracy ±5mm). Pressure sensors: Four sets of distributed pressure sensors are embedded under the track sleepers (two sets for each wheel on each side) to collect the pressure value of a single wheel and the pressure difference between the two sides in real time, and to determine whether the wheel is completely stopped in the designated load-bearing area (an anomaly is triggered if the pressure difference exceeds 10%). Visual recognition equipment: An industrial camera is installed directly in front of the track to capture the relative position of the coupler and the ground marking line (yellow positioning line). The lateral offset is calculated through image recognition algorithms (a centering deviation of ≤3cm for the marking line is acceptable).

[0055] Edge computing processing includes: rule base configuration: the safe operating range is set as a lateral distance of 10-15cm and a longitudinal distance of 5-10cm; the pressure equalization condition is that the pressure difference between the two wheels is ≤5%; and the coupler deviation threshold is ≤3cm. Logical judgment process: the edge computing unit receives three types of data in real time. Only when the distance data is within the safe range, the pressure difference is ≤5%, and the coupler deviation is ≤3cm, does it output a "ready to be in position" signal; if any condition is not met, it marks "position deviation" and displays the specific fault type (such as "longitudinal distance exceeded") on the LED screen.

[0056] In some embodiments, the method of collecting the vehicle number image of the tanker by the image recognition device and converting it into character information includes: deploying multi-angle high-definition cameras on both sides of the oil unloading operation area to dynamically scan the vehicle number areas on the side and end faces of the tanker, and obtaining multiple consecutive images; performing preprocessing such as noise reduction, distortion correction, and image enhancement on the multiple images, using a convolutional neural network character recognition model to segment and recognize the vehicle number characters in the preprocessed image, and generating initial character information; performing string similarity matching between the initial character information and the candidate vehicle numbers of the same batch transportation plan in the refinery incoming oil information database, and when the matching degree exceeds a preset threshold, outputting the finally confirmed vehicle number character information.

[0057] Through multi-angle dynamic scanning, image preprocessing, and intelligent matching algorithms, the accuracy and robustness of vehicle number recognition are improved. The core includes: Multi-angle image acquisition: Deploy high-definition cameras on both sides to cover the vehicle number areas on the side and end faces of the tanker, solving the problem of occlusion in a single view. Multi-level recognition and verification: First, segment and recognize characters through a convolutional neural network (CNN), and then perform string similarity matching with candidate vehicle numbers, double verification ensures the reliability of the recognition result.

[0058] Image acquisition and preprocessing include: Hardware deployment: Install high-definition cameras with pan-tilt heads (frame rate 30fps, resolution 2048×1536) at a height of 3m on both sides of the oil unloading operation area, supporting automatic tracking of the movement of the tanker and dynamically scanning the vehicle number areas (the "tanker number" on the side and the "cab number" on the end face). Preprocessing process: Perform median filtering for noise reduction on 10 consecutive images, correct distortion through perspective transformation (solve the character deformation caused by the shooting angle), and then enhance the contrast with histogram equalization, and output a standardized character area image.

[0059] Character recognition and matching include: Application of the CNN model: Use an improved ResNet+CTC architecture model. After inputting the preprocessed image, it automatically segments individual characters (supporting mixed recognition of Chinese characters, letters, and numbers), and outputs a character sequence and confidence level (such as "Tank A-1234", confidence level 98%). Similarity matching: Calculate the edit distance between the candidate vehicle numbers of the same batch in the incoming oil information database (such as "Tank A-1234" and "Tank B-5678") and the initial recognition result. When the matching degree ≥ 95% (that is, there is at most one character difference), it is confirmed as a valid vehicle number; if the matching degree < 95%, automatically trigger additional shooting by adjacent cameras, and merge the multi-view recognition results for re-verification.

[0060] In some embodiments, the step of initiating metering data acquisition, which involves obtaining oil metering data from the tanker truck using a portable measuring instrument, includes: establishing a wireless communication connection between the portable measuring instrument and the tanker truck's level gauge, temperature sensor, and pressure sensor; automatically synchronizing the clock and verifying the equipment calibration time; sequentially collecting oil level height, temperature, density, and volume data; removing abnormal fluctuation values ​​and taking the average value as valid metering data; and using blockchain technology to hash and store the metering data acquisition time, equipment number, and data content to generate an immutable metering data record.

[0061] Wireless communication enables the linkage of metering equipment, collecting multi-dimensional oil product data, and blockchain technology ensures the immutability of the data. Core features include: Equipment interconnection and data calibration: Portable measuring instruments wirelessly connect to the built-in sensors of tank trucks, synchronizing clocks and verifying equipment calibration status to ensure reliable data timestamps and accuracy. Outlier handling and blockchain notarization: Abnormal fluctuations in the collected data are removed, and key information is uploaded to the blockchain using a hash algorithm to create tamper-proof records.

[0062] Equipment interconnection and data acquisition include: Communication protocol: Portable measuring instruments (such as explosion-proof PDAs) connect to tank truck level gauges (supporting RS485 interface), temperature sensors (Pt100 type), and pressure sensors (accuracy 0.1%FS) via Bluetooth 5.0 or LoRa. Upon initial connection, NTP clocks are automatically synchronized (time error ≤10ms), and the validity period of sensor calibration certificates is verified (<12 months since last calibration). Data acquisition strategy: Continuously collect 10 sets of data on liquid level (unit: mm), temperature (°C), and density (kg / m³). 3 After removing outliers exceeding 2σ (standard deviation), the mean of the remaining data is taken as the valid value (e.g., after removing one outlier level value, the mean of nine values ​​is taken).

[0063] Blockchain-based evidence storage includes: Hash value generation: The collection time (accurate to the second), device number (such as sensor SN code), and data content (liquid level 1234mm, temperature 25.3℃) are combined into a string, and a 256-bit hash value is generated using the SHA-256 algorithm. This hash value is then linked to the hash value of the previous block, forming a chain structure. The evidence storage method involves uploading the hash value and metadata to a consortium blockchain node (jointly maintained by the refinery, railway operator, and regulator). Once the data is uploaded to the blockchain, it cannot be modified, supporting rapid verification of data integrity during subsequent audits using the hash value.

[0064] In some embodiments, the step of comparing and analyzing the collected measurement data with theoretical data from the refinery's incoming oil information database and historical data from the railway tank car measurement database using a preset intelligent algorithm includes: standardizing the theoretical data in the refinery's incoming oil information database to extract the theoretical values ​​and allowable error ranges of oil density and volume; retrieving historical measurement data of the same type of oil for the same car number within the past 12 months from the railway tank car measurement database to calculate the mean, standard deviation, and trend of volume and density; using a dynamic time warping algorithm to calculate the similarity between the current measurement data and the historical data sequence, and constructing a three-dimensional comparison model in conjunction with the allowable error range of the theoretical data; if the similarity is higher than a preset threshold and the data falls within the theoretical error range, the comparison is deemed successful.

[0065] A three-dimensional comparison model is constructed to identify data anomalies through theoretical data normalization, historical data trend analysis, and dynamic time warping algorithms. The core processes include: Data preprocessing: extracting core indicators and error ranges from theoretical data, and retrieving historical data to calculate statistical characteristics (mean, standard deviation, trend). Similarity calculation and model determination: using the dynamic time warping (DTW) algorithm to match the similarity between current data and historical data sequences, combined with a comprehensive determination based on the theoretical error range.

[0066] Data preprocessing includes: Theoretical data processing: Extracting the theoretical density value of the current batch of oil from the refinery's incoming oil information database (e.g., 750 kg / m³). 3 ) and allowable error (±2%), theoretical volume value (e.g., 50m³) 3 The allowable error (±1.5%) forms the theoretical data range [735,765] kg / m³ and [49.25,50.75] m³. 3 Historical data retrieval involves querying the railway tank car metering database for all transportation records of the same type of oil (such as gasoline) within the past 12 months for that car number, extracting the volume and density data for each instance, and calculating historical averages (e.g., average volume 49.8 m³, average density 755 kg / m³). 3 ), standard deviation (volume σ=0.3m) 3 Density σ = 8 kg / m³ 3 ) and trends (e.g., the volume increased by 0.1m³ in the last three instances) 3 ).

[0067] The intelligent comparison algorithm includes: DTW similarity calculation: The current volume / density data sequence (e.g., [50.1, 752]) is time-warped and matched with historical data sequences (e.g., the most recent 3 [49.8, 755], [49.9, 753], [50.0, 754]), and the Euclidean distance similarity is calculated (threshold set to ≥0.95) to reflect whether the data fluctuation pattern conforms to historical patterns. Three-dimensional comparison model: The comparison is considered successful only when the following conditions are met simultaneously: ① Real-time data falls within the theoretical error range; ② The deviation between real-time data and the historical mean is ≤2σ; ③ DTW similarity ≥0.95.

[0068] In some embodiments, the verification based on vehicle number, entry status, and metering data, and the generation of an automatic oil unloading command after successful verification, includes: establishing a verification rule engine and configuring vehicle number uniqueness verification, entry status stability verification, and metering data logical verification rules, including: verifying whether the vehicle number matches the current operation plan in the refinery's incoming oil information database and has not been repeatedly activated; secondly, verifying whether the fluctuation range of the entry status information during the metering data collection period is less than a preset stability threshold; and verifying whether the volume value in the metering data is within a preset data range; when all logical verification rules trigger the pass condition, an encrypted automatic oil unloading command containing a timestamp, vehicle number, and equipment number is generated.

[0069] By establishing a multi-layered verification rule engine that includes vehicle number, location data, and metering data, the security and uniqueness of command generation are ensured. The core features include: Rule engine configuration: defining three types of verification rules—vehicle number uniqueness, location stability, and metering data logic—forming a comprehensive verification system. Encrypted command generation: After successful verification, an encrypted command containing a timestamp and device number is generated to prevent command tampering or repeated execution.

[0070] The rules engine configuration includes: Vehicle number uniqueness verification: Checking whether the currently identified vehicle number is in the oil delivery operation plan list, and whether the unloading instruction for that vehicle number has been activated repeatedly (by querying instructions to generate logs, avoiding duplicate unloading of the same vehicle). Positioning stability verification: Retrieving positioning status data (such as laser ranging values) during the metering data collection period (approximately 2 minutes), calculating the fluctuation range (maximum value - minimum value), lateral / longitudinal distance fluctuation ≤ 5mm and coupler deviation fluctuation ≤ 2cm are considered stable. Metering data logic verification: The volume value must meet the requirement of "0 < real-time volume < tanker truck nominal volume" (e.g., nominal volume 55m³). 3 The real-time volume needs to be between 1 and 55 m³. 3 (between), the density value must conform to the physical properties of the oil (e.g., diesel density > gasoline density).

[0071] Command generation and encryption include: Triggering conditions: When all three rules are met (vehicle number matches and is not duplicated, position is stable, and metering data logic is reasonable), the system automatically generates a command. Command content: Includes a timestamp (accurate to milliseconds), vehicle number, device number (e.g., loading / unloading port ID-001), and command validity period (30 minutes). The command is encrypted using the AES-256 encryption algorithm and a digital signature (generated based on the device's private key) is attached to ensure that the command is immutable and can only be executed once.

[0072] In some embodiments, the railway oil unloading automation control system receives an automatic oil unloading command, confirms the start of the oil unloading status, and starts the oil unloading operation after confirming that the equipment status is normal. This includes: the railway oil unloading automation control system sequentially sends status query commands to the oil unloading pipeline valves, centrifugal pumps, and electrostatic grounding devices, and receives feedback signals from each device; it analyzes the equipment operating parameters, fault codes, and safety interlock status in the feedback signals; if the valve opening feedback value is consistent with the initial state, the centrifugal pump motor temperature is lower than the warning threshold, and the electrostatic grounding resistance is less than the safety limit, then the equipment status is determined to be normal; after the status confirmation is passed, a segmented start command is sent to the field actuator, first opening the oil unloading valve at the bottom of the tank car, then starting the centrifugal pump after a 10-second delay, and simultaneously recording the oil unloading start time and the equipment start log.

[0073] The safe start-up of oil unloading operations is ensured through equipment status queries, safety interlock verification, and a phased start-up strategy. Key aspects include: Full-link equipment status verification: The status of valves, centrifugal pumps, and electrostatic grounding devices are checked sequentially to confirm no faults and effective safety interlocks. Phased start-up control: A delayed start-up strategy of "opening valves first, then starting pumps" is adopted to avoid safety risks caused by sudden changes in pipeline pressure.

[0074] Equipment status query includes: Communication protocol: The automated control system sends a "status query" command to the unloading pipeline valve (electric ball valve) via Modbus TCP protocol to obtain the valve opening degree (0-100%); sends a command to the centrifugal pump control cabinet to obtain the motor temperature (°C) and speed (rpm); and sends a command to the static grounding device to obtain the grounding resistance (Ω, safety limit ≤10Ω).

[0075] Safety interlock verification: The following conditions must be met: ① Valve opening feedback value = 0% (initial state is closed); ② Centrifugal pump motor temperature < 60℃ (early warning threshold); ③ Static grounding resistance ≤ 10Ω; If any condition is not met, the system will return "Equipment abnormal", suspend operation and alarm (e.g., "Static grounding failure").

[0076] The execution steps include: ① Sending a command to open the bottom unloading valve (pneumatic valve) of the tank truck, waiting 5 seconds and then reading the valve feedback signal (100% opening confirmation); ② Delaying for 10 seconds (to ensure no pressure buildup in the pipeline) and then starting the centrifugal pump, recording the start-up time (accurate to the second); Log recording: Real-time recording of the sending time of each command, equipment response time, and feedback parameters to form the "Unloading Start-up Log" for subsequent fault tracing (e.g., valve response timeout > 15 seconds is marked as equipment failure).

[0077] In some embodiments, after the oil unloading is completed, a final status confirmation is performed, and an oil receipt and dispatch certificate information report is generated. This includes: real-time monitoring of liquid level changes using tank truck level sensors; determining that the oil unloading operation is complete when the liquid level data remains below the safe low threshold for 5 minutes; the control system automatically shuts down the centrifugal pump and oil unloading valve, and collects status data after equipment reset, including valve closure feedback signals, pipeline pressure zeroing values, and empty truck status images captured by on-site video; integrating the truck number, operation time, metering data, equipment status records, and blockchain evidence hash values ​​into a preset report template, automatically generating a PDF format oil receipt and dispatch certificate information report, pushing it to the refinery information management system through an encrypted channel, and simultaneously backing it up to a distributed database.

[0078] The system continuously monitors the liquid level to determine when unloading is complete, automatically resets the equipment, and generates an electronic report containing blockchain-based evidence. Key features include: Intelligent operation completion determination: using a continuously decreasing liquid level below a safety threshold as the termination condition to avoid human error in determining residual levels; and end-to-end data archiving: integrating vehicle number, metering data, equipment status, blockchain hash values, and other information to generate standardized reports and transmit them encrypted.

[0079] The completion status confirmation includes: Liquid level monitoring: Data is collected in real time through the liquid level sensor built into the tank truck. When the liquid level value is less than 50mm for 5 consecutive minutes (safe low threshold, corresponding to a remaining volume of less than 0.5m³), the liquid level is checked. 3 When the oil unloading is completed (to prevent misjudgment due to liquid level fluctuations), the equipment is reset: the centrifugal pumps are shut down in sequence (stop the pumps first, and then close the pipeline valves after a 5-second delay), and the status after reset is collected: valve feedback opening 0%, pipeline pressure 0MPa, static grounding device signal disconnection, and the on-site camera is triggered to capture the empty vehicle status (to verify that there is no leakage at the bottom of the tanker).

[0080] Report generation and transmission include: Report content: including vehicle number, operation time (start / end), and metering data (original 50m). 3 49.8m were actually received. 3The data includes: loss rate (0.4%), equipment number, blockchain notarized hash value (linked to metering data block), and any abnormal records. Generation and transmission: Data is automatically populated using a template engine (such as Freemarker), generating a PDF report and encrypting it (password generated from vehicle number + timestamp). The report is pushed to the refinery's ERP system via HTTPS, and the report hash value is simultaneously stored on the blockchain. A distributed database (such as HBase) backs up the entire dataset, supporting fast retrieval by vehicle number (retrieval latency ≤ 2 seconds).

[0081] In some embodiments, such as Figure 2 As shown, the business logic corresponding to the provided method includes: start, filling in oil arrival data registration, oil tanker entry identification, oil tanker number identification and comparison, comparison pass (with refinery oil arrival information database), pass (manual correction if not pass), manual confirmation, metering data collection (via portable measuring instrument), metering data comparison and analysis, comparison pass (with refinery oil arrival information database and railway tanker metering database), pass (abnormal situation registration if not pass), verification and confirmation, implementation of automatic oil unloading, confirmation of start and end status of oil unloading operation by railway oil unloading automation control system, automatic generation of report (oil receiving and dispatching certificate information report), end.

[0082] In some embodiments, a Long Short-Term Memory (LSTM) network is trained using historical equipment operating data to construct a fault prediction model for key equipment such as centrifugal pumps and valves, enabling preventative maintenance and reducing the risk of unplanned downtime. Core technologies include: Multi-dimensional feature engineering: extracting time-series features such as equipment vibration frequency, temperature change rate, and current fluctuations, and combining them with work order records to construct a fault label dataset. LSTM anomaly detection model: training a baseline model using normal equipment operating data, and identifying early fault signs through residual analysis of real-time data and predicted values.

[0083] Data Acquisition and Feature Construction: A triaxial accelerometer (sampling frequency 1000Hz) was installed in the centrifugal pump bearing housing, and a current transformer was installed in the motor junction box. Vibration waveforms (time-domain mean, kurtosis), RMS current values, and temperature gradients (ΔT / 10min) were collected every 5 minutes, representing 12 dimensions of features. Historical fault data (such as bearing wear and impeller corrosion) were labeled, and a balanced dataset with a positive-to-negative sample ratio of 1:3 was constructed. Data from the 72 hours prior to the fault was marked as early warning samples.

[0084] Model training and application include: Training process: A 2-layer LSTM network (128 neurons per layer) is used. The input is the feature sequence from the previous 24 hours, and the output is the probability of failure in the next 6 hours (threshold set to ≥85% to trigger an early warning). The Adam optimizer is used, with the loss function being binary cross-entropy, and the validation set accuracy is ≥92%. Real-time prediction: The control system inputs the current feature sequence to the model every 10 minutes. If the predicted failure probability is >80%, a maintenance work order (including the failure type and suggested repair time) is automatically generated, and a backup pump is scheduled for switching. Simultaneously, an early warning is pushed to the maintenance terminal (response latency ≤30 seconds).

[0085] In some embodiments, to address the challenge of license plate recognition under different lighting and dirt occlusion scenarios, transfer learning is employed to optimize the CNN model, combined with domain adaptation algorithms to improve generalization ability. Core technologies include: Cross-domain data augmentation: CycleGAN is used to generate simulated images such as those from rainy days, nighttime scenes, and oil stain coverage to expand the training dataset. Meta-learning fine-tuning mechanism: Based on the pre-trained model, the classification layer is rapidly fine-tuned for new scenarios to address the problem of decreased recognition accuracy in small sample scenarios.

[0086] Data Augmentation and Model Architecture: Simulated Data Generation: Utilizing CycleGAN to learn the mapping relationship between normal license plate images and images of adverse scenes, 50,000 augmented data images (e.g., adding Gaussian noise, rain streaks, and oil stain masks) were generated and mixed with real scene data (20,000 images) for training. Transfer Learning Framework: Based on ResNet50, the first four convolutional layers were frozen, and the last three layers were replaced with adaptive feature extraction layers, supporting feature normalization for different lighting / occlusion scenarios.

[0087] Real-time recognition optimization includes: Scene detection: Determining the current scene type (e.g., "low light at night" or "oil stain obstruction") based on prior features such as image brightness, contrast, and the proportion of stained areas, and automatically loading the corresponding fine-tuned model (switching latency ≤200ms). Dynamic fusion strategy: A voting decision is made based on the recognition results from multiple cameras. If the confidence level of a single camera in a certain scene is <90%, image fusion of adjacent cameras is triggered (e.g., reshooting from the left camera + image enhancement from the right camera), and the fused license plate number is output (recognition accuracy improved to 99.2%).

[0088] In some embodiments, a reinforcement learning model for oil unloading operation scheduling is constructed to dynamically optimize the tanker unloading sequence and equipment allocation strategy with the goal of minimizing the total operation time. Core technologies include: State-space modeling: Encoding parameters such as tanker type (light oil / heavy oil), equipment availability, and metering data integrity into state vectors. Action-space design: Defining 12 scheduling actions such as "allocate to loading / unloading port 1" and "prioritize unloading heavy oil tankers," and designing a reward function that combines waiting time and equipment load balancing.

[0089] Environmental Modeling and Reward Function: State Vector: Contains 20 parameters including the queue of vehicles to be unloaded (vehicle number, oil type, volume), the current status of each loading / unloading port equipment (valve / pump availability, last maintenance time), and storage area capacity limits. Reward Function: +100 for completing the unloading of one vehicle, -50 for equipment load exceeding 80%, and -10 for every minute of delay. The policy network is trained using the Proximal Policy Optimization (PPO) algorithm.

[0090] Real-time scheduling application: Input interface: Obtain the list of trucks to be unloaded (including estimated arrival time and priority) in real time from the refinery's oil arrival information database. A scheduling re-optimization is triggered every 3 newly added trucks. Decision output: The model outputs the optimal scheduling sequence (e.g., "Truck A → Loading / Unloading Port 2, Truck B → Loading / Unloading Port 1 (Priority Heavy Oil)"), and simultaneously generates an equipment preheating plan (starting the centrifugal pump 5 minutes in advance to reduce idling losses). After manual confirmation, the plan is automatically executed, reducing the average scheduling time by 30%.

[0091] In some embodiments, a spatiotemporal graph neural network (ST-GNN) incorporating equipment status, personnel location, and video stream is constructed to identify complex abnormal behaviors such as personnel misoperation and abnormal equipment linkage in the unloading area. Core technologies include: Multimodal data graph modeling: abstracting personnel trajectories detected by cameras, equipment sensor data, and operational process nodes into graph nodes, with edges representing spatiotemporal relationships. Anomaly score calculation: learning normal behavior patterns through a graph convolutional neural network (GCN), and determining subgraph structures with a deviation > 3σ as abnormal.

[0092] Graph structure construction and data fusion include: node definition: personnel nodes (ID, location, action tag), equipment nodes (number, status parameters), and process nodes (work steps, timestamp), totaling 3 categories and 15 types of node attributes.

[0093] Edge definition: When the distance between personnel and equipment is less than 2m, an interactive edge is established. When the equipment status changes and process nodes are established, a time sequence edge is established to form a dynamic spatiotemporal graph (updated every 10 seconds).

[0094] The anomaly detection process includes: Model training: Training ST-GNN using normal operating data to learn the node state transition rules (such as the strong correlation between "opening the valve → starting the pump after 10 seconds").

[0095] Real-time detection includes: when "pump starts without confirmation of entry" (the equipment starts before the personnel have completed the confirmation process) or "multiple people enter the danger zone at the same time" (exceeding the safe number of people limit), the graph model calculates an anomaly score > 0.85, immediately triggers an audible and visual alarm and freezes the equipment control authority, and pushes an alarm information containing the anomaly subgraph to the safety management platform (response time ≤ 1 second).

[0096] In some embodiments, to address the data privacy protection needs of refineries and railways, a federated learning framework is used to collaboratively train a metrological data verification model, improving comparison accuracy without sharing the original data. Core technologies include: a horizontal federated learning architecture: the refinery (theoretical data) and the railway (historical metrological data) act as two data providers, collaboratively training a neural network, locally calculating gradients, and encrypting and uploading them for aggregation. Differential privacy protection: Laplace noise is added during gradient aggregation to ensure data privacy compliance (ε=0.5).

[0097] The federated learning process design includes: Data preprocessing: The refinery prepares theoretical data (oil density, volume range), and the railway prepares historical measurement data (vehicle number, measured value, timestamp). The data from both sides are aligned by vehicle number ID, and 10% of the common IDs are reserved for model validation. The model architecture adopts a two-layer fully connected neural network (10-dimensional input layer, 32-dimensional hidden layer, and 1-dimensional comparison result output layer). The refinery inputs the theoretical error range, and the railway inputs the historical mean / standard deviation. Both sides train locally and then upload the gradients.

[0098] Collaborative verification applications include: Model updates: Federated training iterations are performed weekly, and after ≥5 aggregations, the model comparison accuracy stabilizes above 98%. Real-time verification: When new metrological data is generated, the refinery inputs theoretical parameters, and the railway inputs historical features. Both parties call the federated model for encrypted comparison (using homomorphic encryption technology to protect intermediate results), and return the verification result within 10 seconds. This protects the privacy of commercial data and solves the problem of insufficient generalization ability of models from single data sources.

[0099] In some embodiments, a digital twin of the oil unloading area is constructed, and virtual scene simulation is driven by real-time data. This is combined with physical models and machine learning to predict pressure fluctuations and leakage risks during the oil unloading process. Core technologies include: Multiphysics coupling modeling: establishing a fusion simulation framework based on the Navier-Stokes equations for oil flow, equipment thermodynamic models, and sensor data. Twin data-driven approach: calibrating virtual model parameters using real-time liquid level and pressure data, and predicting the process status for the next 30 minutes through simulation.

[0100] The construction of the digital twin includes: Geometric modeling: using BIM technology to build a 3D model of the unloading area (accuracy ±5mm), integrating CAD drawings of tank trucks, pipelines, and pumps, and defining material properties (such as pipeline thermal conductivity and oil viscosity). Data interface: acquiring 100 sets of sensor data per second (liquid level, pressure, temperature), fusing them through Kalman filtering to drive dynamic updates of the virtual scene, with errors controlled within 1.5%.

[0101] Risk prediction applications include: Pressure fluctuation prediction: When the centrifugal pump speed changes, the twin model simulates the pipeline pressure distribution through computational fluid dynamics (CFD). If the predicted pressure at a certain bend exceeds the design threshold (1.2 MPa), a "pipeline overload risk" warning is issued 5 minutes in advance, and the pump speed is automatically adjusted to a safe range (adjustment step ≤ 50 rpm). Leakage simulation: A "valve seal aging" fault is injected into the virtual scenario, simulating the leakage diffusion path. Combined with wind direction and terrain data, the impact range is predicted, guiding the optimization of on-site emergency drill plans (such as determining a 30-meter safety warning radius), achieving closed-loop management of "prediction-prevention-response".

[0102] Please see Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of the intelligent verification system 200 for railway tank cars provided in this application embodiment. The intelligent verification system 200 for railway tank cars is used to execute the steps of the intelligent verification method for railway tank cars shown in the above embodiments. The intelligent verification system 200 for railway tank cars can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0103] like Figure 3 As shown, the intelligent verification system 200 for railway tank cars includes: The data acquisition unit 201 is used to acquire oil-related data, obtain the tanker truck's entry status information through the recognition device, and collect the tanker truck's license plate number image and convert it into character information based on the image recognition device. The information comparison unit 202 is used to compare the character information with the tank truck number information of the corresponding batch in the refinery's oil arrival information database. If the comparison is successful, the metering data acquisition is started, and the metering data of the oil in the tank truck is obtained through a portable measuring instrument. The instruction generation unit 203 is used to compare and analyze the collected metering data with the theoretical data of the refinery's oil arrival information database and the historical data of the railway tank car metering database using a preset intelligent algorithm. If the comparison is successful, the system verifies the data based on the car number, the entry status, and the metering data. Once the verification is successful, an automatic oil unloading instruction is generated. The intelligent verification unit 204 is used by the railway oil unloading automation control system to receive automatic oil unloading instructions, confirm the start status of oil unloading, start the oil unloading operation after confirming that the equipment status is normal, confirm the end status after the oil unloading is completed, generate an oil receipt and dispatch certificate information report, and complete the intelligent verification of railway tank cars.

[0104] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the railway tank car intelligent verification system and its modules described above can be referred to the corresponding processes in the embodiments of the railway tank car intelligent verification method described above, and will not be repeated here.

[0105] The aforementioned intelligent verification method for railway tank cars can be implemented as a computer program, which can be used in various ways, such as... Figure 3 It runs on the system shown.

[0106] Please see Figure 4 , Figure 4 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0107] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any intelligent verification method for railway tank cars.

[0108] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0109] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any intelligent verification method for railway tank cars.

[0110] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0111] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0112] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: After obtaining relevant data on incoming oil, the system uses identification equipment to obtain the positioning status information of the tanker truck, and uses image recognition equipment to collect the vehicle number image of the tanker truck and convert it into character information. Compare the character information with the tanker truck number information of the corresponding batch in the refinery's oil arrival information database: if the comparison is successful, start the metering data acquisition and obtain the oil metering data in the tanker truck through a portable measuring instrument; The collected metering data is compared and analyzed with theoretical data from the refinery's oil arrival information database and historical data from the railway tank car metering database using a preset intelligent algorithm. If the comparison passes, the data is verified based on the car number, entry status, and metering data. Once the verification passes, an automatic oil unloading command is generated. The railway oil unloading automation control system receives the automatic oil unloading command, confirms the start of the oil unloading status, starts the oil unloading operation after confirming that the equipment status is normal, confirms the end status after the oil unloading is completed, generates an oil receipt and dispatch certificate information report, and completes the intelligent verification of railway tank cars.

[0113] In some embodiments, obtaining the positioning status information of the oil tanker through the identification device includes: deploying a laser ranging sensor, a pressure sensor, and a visual recognition device at a preset position on the track in the railway unloading operation area; collecting distance data between the oil tanker and the loading / unloading port through the laser ranging sensor; collecting wheel-to-track pressure distribution data through the pressure sensor; and collecting the relative position image of the coupler and the track marking line through the visual recognition device; inputting the distance data, pressure distribution data, and position image into an edge computing unit; performing logical judgment based on a preset positioning rule library; and determining that the oil tanker is ready to be positioned if the distance data is within a safe operating range, the pressure distribution data meets the track load balance condition, and the deviation between the coupler and the marking line in the position image is less than a preset threshold.

[0114] In some embodiments, the step of acquiring the tanker truck's license plate image and converting it into character information using an image recognition device includes: deploying multi-angle high-definition cameras on both sides of the unloading area to dynamically scan the license plate area on the side and end of the tanker truck to acquire multiple consecutive images; performing noise reduction, distortion correction, and image enhancement preprocessing on the multiple images; using a convolutional neural network character recognition model to segment and recognize the license plate characters in the preprocessed images to generate initial character information; performing string similarity matching between the initial character information and candidate license plates from the same batch of transportation plans in the refinery's oil arrival information database; and outputting the finally confirmed license plate character information when the matching degree exceeds a preset threshold.

[0115] In some embodiments, the step of initiating metering data acquisition, which involves obtaining oil metering data from the tanker truck using a portable measuring instrument, includes: establishing a wireless communication connection between the portable measuring instrument and the tanker truck's level gauge, temperature sensor, and pressure sensor; automatically synchronizing the clock and verifying the equipment calibration time; sequentially collecting oil level height, temperature, density, and volume data; removing abnormal fluctuation values ​​and taking the average value as valid metering data; and using blockchain technology to hash and store the metering data acquisition time, equipment number, and data content to generate an immutable metering data record.

[0116] In some embodiments, the step of comparing and analyzing the collected measurement data with theoretical data from the refinery's incoming oil information database and historical data from the railway tank car measurement database using a preset intelligent algorithm includes: standardizing the theoretical data in the refinery's incoming oil information database to extract the theoretical values ​​and allowable error ranges of oil density and volume; retrieving historical measurement data of the same type of oil for the same car number within the past 12 months from the railway tank car measurement database to calculate the mean, standard deviation, and trend of volume and density; using a dynamic time warping algorithm to calculate the similarity between the current measurement data and the historical data sequence, and constructing a three-dimensional comparison model in conjunction with the allowable error range of the theoretical data; if the similarity is higher than a preset threshold and the data falls within the theoretical error range, the comparison is deemed successful.

[0117] In some embodiments, the verification based on vehicle number, entry status, and metering data, and the generation of an automatic oil unloading command after successful verification, includes: establishing a verification rule engine and configuring vehicle number uniqueness verification, entry status stability verification, and metering data logical verification rules, including: verifying whether the vehicle number matches the current operation plan in the refinery's incoming oil information database and has not been repeatedly activated; secondly, verifying whether the fluctuation range of the entry status information during the metering data collection period is less than a preset stability threshold; and verifying whether the volume value in the metering data is within a preset data range; when all logical verification rules trigger the pass condition, an encrypted automatic oil unloading command containing a timestamp, vehicle number, and equipment number is generated.

[0118] In some embodiments, the railway oil unloading automation control system receives an automatic oil unloading command, confirms the start of the oil unloading status, and starts the oil unloading operation after confirming that the equipment status is normal. This includes: the railway oil unloading automation control system sequentially sends status query commands to the oil unloading pipeline valves, centrifugal pumps, and electrostatic grounding devices, and receives feedback signals from each device; it analyzes the equipment operating parameters, fault codes, and safety interlock status in the feedback signals; if the valve opening feedback value is consistent with the initial state, the centrifugal pump motor temperature is lower than the warning threshold, and the electrostatic grounding resistance is less than the safety limit, then the equipment status is determined to be normal; after the status confirmation is passed, a segmented start command is sent to the field actuator, first opening the oil unloading valve at the bottom of the tank car, then starting the centrifugal pump after a 10-second delay, and simultaneously recording the oil unloading start time and the equipment start log.

[0119] In some embodiments, after the oil unloading is completed, a final status confirmation is performed, and an oil receipt and dispatch certificate information report is generated. This includes: real-time monitoring of liquid level changes using tank truck level sensors; determining that the oil unloading operation is complete when the liquid level data remains below the safe low threshold for 5 minutes; the control system automatically shuts down the centrifugal pump and oil unloading valve, and collects status data after equipment reset, including valve closure feedback signals, pipeline pressure zeroing values, and empty truck status images captured by on-site video; integrating the truck number, operation time, metering data, equipment status records, and blockchain evidence hash values ​​into a preset report template, automatically generating a PDF format oil receipt and dispatch certificate information report, pushing it to the refinery information management system through an encrypted channel, and simultaneously backing it up to a distributed database.

[0120] The embodiments of this application also provide a computer-readable storage medium storing a computer program, the computer program including program instructions, and the processor executing the program instructions to implement the steps of the intelligent verification method for railway tank cars provided in the above embodiments of this application.

[0121] The computer-readable storage medium can be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium can also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0122] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for intelligent verification of railway tank cars, characterized in that, include: After acquiring relevant oil arrival data, the positioning status information of the oil tanker is obtained through identification equipment. This includes: deploying laser rangefinders, pressure sensors, and visual recognition equipment at preset positions on the track in the railway unloading area; collecting distance data between the oil tanker and the loading / unloading port through the laser rangefinders; collecting wheel-to-track pressure distribution data through the pressure sensors; and collecting relative position images of the coupler and track marking lines through the visual recognition equipment. The distance data, pressure distribution data, and position images are input into an edge computing unit, which performs logical judgment based on a preset positioning rule library. If the distance data is within the safe operating range, the pressure distribution data meets the track load balance condition, and the position image shows a deviation between the coupler and the marking line... If the value is less than a preset threshold, the tanker truck's positioning status is determined to be ready. The process involves: acquiring images of the tanker truck's license plate number using image recognition equipment and converting them into character information; deploying multi-angle high-definition cameras on both sides of the unloading area to dynamically scan the license plate number area on the sides and ends of the tanker truck, acquiring multiple consecutive images; performing noise reduction, distortion correction, and image enhancement preprocessing on the multiple consecutive images; using a convolutional neural network character recognition model to segment and recognize the license plate number characters in the preprocessed images to generate initial character information; matching the initial character information with the candidate license plate numbers of the same batch of transport plans in the refinery's oil arrival information database based on string similarity; and outputting the final confirmed license plate number character information when the matching degree exceeds a preset threshold. Compare the character information with the tanker truck number information of the corresponding batch in the refinery's oil arrival information database: if the comparison is successful, start the metering data acquisition and obtain the oil metering data in the tanker truck through a portable measuring instrument; The collected metering data is compared and analyzed with theoretical data from the refinery's oil arrival information database and historical data from the railway tank car metering database using a preset intelligent algorithm. If the comparison passes, the data is verified based on the car number, entry status, and metering data. Once the verification passes, an automatic oil unloading command is generated. The railway oil unloading automation control system receives automatic oil unloading commands, confirms the start of unloading status, and initiates the unloading operation after confirming that the equipment is in normal condition. After unloading is completed, it confirms the end status and generates an oil receipt and dispatch certificate information report, including: real-time monitoring of liquid level changes through tank car level sensors; when the liquid level data remains below the safe low threshold for 5 minutes, the unloading operation is determined to be completed; the control system automatically shuts down the centrifugal pump and unloading valve, and collects status data after equipment reset, including valve closure feedback signals, pipeline pressure zeroing values, and empty car status images captured by on-site video; it integrates car number, operation time, metering data, equipment status records, and blockchain evidence hash values ​​into a preset report template, automatically generates a PDF format oil receipt and dispatch certificate information report, pushes it to the refinery information management system through an encrypted channel, and simultaneously backs it up to the distributed database to complete intelligent verification of railway tank cars.

2. The method according to claim 1, characterized in that, The initiation of metering data acquisition involves obtaining metering data of the oil in the tanker truck using a portable measuring instrument, including: The portable measuring instrument establishes a wireless communication connection with the tank truck's level gauge, temperature sensor, and pressure sensor, automatically synchronizes the clock, and verifies the equipment calibration time; it sequentially collects oil level height, temperature, density, and volume data, and takes the average value after removing abnormal fluctuation values ​​as valid measurement data; By using blockchain technology, the collection time, equipment number, and data content of metering data are stored using hash values, generating unalterable metering data records.

3. The method according to claim 1, characterized in that, The process involves comparing and analyzing the collected metering data with theoretical data from the refinery's oil supply information database and historical data from the railway tank car metering database using a preset intelligent algorithm, including: The theoretical data in the refinery's incoming oil information database is standardized to extract the theoretical values ​​and allowable error ranges of oil density and volume; historical measurement data of the same type of oil for the same car number within the past 12 months are retrieved from the railway tank car measurement database to calculate the mean, standard deviation, and trend of volume and density. The similarity between the current measurement data and the historical data sequence is calculated using a dynamic time warping algorithm. A three-dimensional comparison model is constructed by combining the theoretical error range of the data. If the similarity is higher than the preset threshold and the data falls within the theoretical error range, the comparison is deemed to have passed.

4. The method according to claim 1, characterized in that, The process involves verification based on vehicle number, parking status, and metering data. Upon successful verification, an automatic oil unloading command is generated, including: Establish a verification rule engine and configure rules for vehicle number uniqueness verification, entry status stability verification, and metering data logic verification, including: verifying whether the vehicle number matches the current operation plan in the refinery's oil arrival information database and has not been activated repeatedly; secondly, verifying whether the fluctuation range of the entry status information during the metering data collection period is less than the preset stability threshold; and verifying whether the volume value in the metering data is within the preset data range. When all logical verification rules trigger the pass condition, an encrypted automatic oil unloading instruction containing a timestamp, vehicle number, and device number is generated.

5. The method according to claim 1, characterized in that, The railway oil unloading automation control system receives the automatic oil unloading command, confirms the start of the oil unloading status, and starts the oil unloading operation after confirming that the equipment status is normal, including: The railway oil unloading automation control system sequentially sends status query commands to the oil unloading pipeline valves, centrifugal pumps, and electrostatic grounding devices, and receives feedback signals from each device. Analyze the equipment operating parameters, fault codes and safety interlock status in the feedback signal. If the valve opening feedback value is consistent with the initial state, the centrifugal pump motor temperature is lower than the warning threshold and the electrostatic grounding resistance is less than the safety limit, then the equipment status is determined to be normal. After the status is confirmed, a segmented start command is sent to the on-site actuator. First, the unloading valve at the bottom of the tanker is opened, and after a 10-second delay, the centrifugal pump is started. The unloading start time and equipment start log are recorded simultaneously.

6. A railway tank car intelligent verification system, used to implement the method as described in any one of claims 1-5, characterized in that, include: The data acquisition unit is used to acquire relevant data on incoming oil, obtain the positioning status information of the oil tanker through the recognition device, and collect the vehicle number image of the oil tanker and convert it into character information according to the image recognition device. The information comparison unit is used to compare the character information with the tanker truck number information of the corresponding batch in the refinery's oil arrival information database. If the comparison is successful, the metering data acquisition is started, and the metering data of the oil in the tanker is obtained through a portable measuring instrument. The instruction generation unit is used to compare and analyze the collected metering data with the theoretical data of the refinery's oil arrival information database and the historical data of the railway tank car metering database using a preset intelligent algorithm. If the comparison is successful, the unit verifies the data based on the car number, the entry status, and the metering data. Once the verification is successful, an automatic oil unloading instruction is generated. The intelligent verification unit is used by the railway oil unloading automation control system to receive automatic oil unloading commands, confirm the start status of oil unloading, start the oil unloading operation after confirming that the equipment status is normal, confirm the end status after oil unloading is completed, generate an oil receipt and dispatch certificate information report, and complete the intelligent verification of railway tank cars.

7. A computer device, characterized in that, The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the method as described in any one of claims 1 to 5.

Images